Abstract 392P
Background
Previous studies on liquid biopsy-based detection of advanced colorectal adenoma (advCRA) and early stage adenocarcinoma (CRC) were limited by low sensitivity. To this end, we performed a prospective study and established an integrated model using multiple fragmentomic profiles of plasma cell-free DNA (cfDNA) for accurately and cost-effectively detecting stage 0/I CRC and advCRA.
Methods
This study enrolled a total of 621 participants, including 298 CRC patients (34 stage 0, 264 stage I), 92 advCRA patients and 231 healthy controls. Plasma cfDNA samples were prepared for low coverage whole-genome sequencing (∼5X). Participants were randomly divided into a training cohort (N = 310) and a test cohort (N = 311). An ensemble stacked model differentiating healthy controls from advCRA/CRC patients was trained using four machine learning models and five cfDNA fragmentomic features, including fragment size distribution and ratio, end and breakpoint motif, and copy number alteration, which was then validated in the test cohort.
Results
Our model showed an Area Under the Curve (AUC) of 0.988 for differentiating advCRA/CRC patients from healthy individuals. The model performed even better for identifying CRC patients (AUC 0.990) compared to advCRA patients (AUC 0.982). At 94.8% specificity, the sensitivities for detecting advCRA and CRC reached 95.7% and 98.0% (stage 0: 94.1%; stage I: 98.5%), respectively. Promisingly, the detection sensitivity has reached 100% and 97.6% in CRC patients with negative fecal occult or CEA blood test results, respectively. In advCRA subgroup, our model showed high sensitivities for detecting different grades of dysplasia (high: 91.3%, low: 100%) and for pedunculated (92.9%) or sessile adenoma (96.9%). Finally, our model maintained promising performances (mean AUC: 0.985, 92.6% sensitivity at 94.8% specificity) even when sequencing depth was downsampled to 1X.
Conclusions
Our integrated predictive model using plasma cfDNA fragmentomic profiles demonstrated an unprecedented detection sensitivity for advCRA and very early-stage CRC, shedding light on more accurate non-invasive colorectal cancer screening in clinical practice.
Clinical trial identification
Editorial acknowledgement
Legal entity responsible for the study
Fudan University Shanghai Cancer Center.
Funding
Fudan University Shanghai Cancer Center.
Disclosure
W. Tang: Financial Interests, Personal, Full or part-time Employment: Nanjing Geneseeq Technology Inc. H. Bao: Financial Interests, Personal, Full or part-time Employment: Nanjing Geneseeq Technology Inc. R. Liu: Financial Interests, Personal, Full or part-time Employment: Nanjing Geneseeq Technology Inc. X. Chen: Financial Interests, Personal, Full or part-time Employment: Nanjing Geneseeq Technology Inc. S. Wu: Financial Interests, Personal, Full or part-time Employment: Nanjing Geneseeq Technology Inc. X. Wu: Financial Interests, Personal, Full or part-time Employment: Nanjing Geneseeq Technology Inc. Y. Shao: Financial Interests, Personal, Member of the Board of Directors: Nanjing Geneseeq Technology Inc. All other authors have declared no conflicts of interest.